A First Step Towards Autonomic Optical Burst Switched Networks

In recent years, Optical Burst Switching (OBS) is emerging as the dominant switching paradigm for transmitting IP traffic over WDM optical networks. This is mainly because OBS effectively combines the advantages of the two other switching paradigms for WDM networks, namely circuit and packet switching. The large volume of data combined with the size and complexity of communication networks, make human management of such systems under network failures and other environment changes ill-affordable, and necessitate autonomy. In this paper, we propose for the first time, a self-optimizing and self-protecting autonomic OBS network system, which uses Learning Automata to learn and adapt to the state of the network. This self-awareness aids the system in making intelligent decisions. In this paper, we develop solutions for three fundamental problems: (i) routing, (ii) wavelength selection, and (iii) protection